Li Xiaoyang wrote: > I am newbie in Linux world. I know that the Linux philosophy is that > executing a command just as it is designated without any verbose > information.
Small is beautiful Simple is better than completeness Silence is golden Those are just a few items that come to my mind with your statement. There are a number of software philosophies that come into play with this type of design. You might want to browse the net. I found these on wikipedia. http://en.wikipedia.org/wiki/Unix_philosophy http://en.wikipedia.org/wiki/List_of_software_development_philosophies http://en.wikipedia.org/wiki/UNIX-HATERS_Handbook > I also know that the root user has too strong powers to change every > detail of the system. However, when it is connected with the most > essential system files, we should give a warning. What about when > "sudo rm -rf /bin", it gives a conformation or asks the user to > input something, and when "sudo -rff /bin", it will do the command > with saying anything? How would you define that list? How would *I* define that list? Will those two lists ever converge to a meaningful set acceptable to both? Probably not. The normal way to deal with this is to alias 'rm' to 'rm -i' so that rm is always in interactive mode. This allows people who wish that level of interactivity to have it while allowing others who do not to avoid it. You are actually proposing much more than simply an option to rm. You are actually asking questions about the core fundamental principles and philosophies of system design and human interfaces. The one in coreutils has been in place for 30 years. Sure questions are good. But changing such long existing behavior is not done lightly. Bob _______________________________________________ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
